In the context of Empirical Risk Minimization, which of the following scenarios is most likely to lead to underfitting while impacting the generalization error negatively?

Question

Seekh · Accepted Answer

Empirical Risk Minimization chooses a hypothesis that minimizes training error, but if the hypothesis class is too simple—such as a linear model for data that follows a curved pattern—then the model will be unable to capture the underlying relationships, producing high bias. Because the model cannot fit the training data well, its training error remains large and its generalization error also stays high. This situation is classic underfitting, where the model is too restrictive and fails to learn the true pattern. For example, fitting a straight line to points that actually follow a quadratic curve will leave large residuals both on the training set and on new data.

In the context of Empirical Risk Minimization, which of the following scenarios is most likely to lead to underfitting while impacting the generalization error negatively?

Learning Path

Choose the Best Answer

Understanding the Answer

Key Concepts

Deep Dive: Empirical Risk Minimization

Definition

Topic Definition

Ready to Master More Topics?